UniGPT: benchmark results

Provider: Other.

Aggregate Rating 1509 ± 23, rank #2211 of 10916 rated models, from 66 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenCompass LLM - Language (CompassBench 2404)60.9Score (%)100
OpenCompass LLM - Language - Chinese (CompassBench 2404)64.7Score (%)100
OpenCompass LLM - Reasoning - Chinese (CompassBench 2404)56.6Score (%)100
OpenCompass Language - Intention Recognition - Chinese (CompassBench 2404)84Score (%)100
OpenCompass Language - Traditional Cultural Understanding - Chinese (CompassBench 2404)73Score (%)100
OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2404)60.5Score (%)100
OpenCompass Language - Chinese Semantic Understanding - Chinese (CompassBench 2404)75Score (%)96.7
OpenCompass Reasoning - Common Sense - English (CompassBench 2404)34.5Score (%)96.7
OpenCompass Reasoning - Inductive - Chinese (CompassBench 2404)36.3Score (%)96.7
OpenCompass Reasoning - Abductive - English (CompassBench 2404)90Score (%)94.8
OpenCompass LLM - Knowledge - English (CompassBench 2404)77Score (%)93.3
OpenCompass LLM - Language - English (CompassBench 2404)57.1Score (%)93.3

Interactive version: theaggregate.ai/model?slug=unigpt · How It Works · Data refreshed daily, snapshot 2026-11-08.